环境

etcd-4节点上拷贝配置文件和启动命令，暂不启动（etcd-4节点上操作）
将etcd-4故障节点剔除集群（etcd-1节点上操作）

一、下线故障etcd-4节点

命令行下线etcd4
下线节点

#list 查看节点状态
[root@k8s-master1 etcd]# ETCDCTL_API=3 /opt/etcd/bin/etcdctl --cacert=/opt/etcd/ssl/ca.pem --cert=/opt/etcd/ssl/server.pem --key=/opt/etcd/ssl/server-key.pem --endpoints="https://10.0.19.128:2379" member list --write-out=table
+------------------+---------+--------+--------------------------+--------------------------+
|        ID        | STATUS  |  NAME  |        PEER ADDRS        |       CLIENT ADDRS       |
+------------------+---------+--------+--------------------------+--------------------------+
|   db5a2c24a8bf7a | started | etcd-2 | https://10.0.19.129:2380 | https://10.0.19.129:2379 |
| 318ff80236db7a1e | started | etcd-1 | https://10.0.19.127:2380 | https://10.0.19.127:2379 |
| ea1f30f33a754dac | started | etcd-4 | https://10.0.19.128:2380 | https://10.0.19.128:2379 |
| ee6138446f578296 | started | etcd-3 | https://10.0.19.130:2380 | https://10.0.19.130:2379 |
+------------------+---------+--------+--------------------------+--------------------------+
#delete 删除etcd-4节点
[root@k8s-master1 etcd]# ETCDCTL_API=3 /opt/etcd/bin/etcdctl --cacert=/opt/etcd/ssl/ca.pem --cert=/opt/etcd/ssl/server.pem --key=/opt/etcd/ssl/server-key.pem --endpoints="https://10.0.19.127:2379" member remove ea1f30f33a754dac
Member ea1f30f33a754dac removed from cluster 983f3fe4681bf439
#list 查看节点状态
[root@k8s-master1 etcd]# ETCDCTL_API=3 /opt/etcd/bin/etcdctl --cacert=/opt/etcd/ssl/ca.pem --cert=/opt/etcd/ssl/server.pem --key=/opt/etcd/ssl/server-key.pem --endpoints="https://10.0.19.127:2379" member list --write-out=table
+------------------+---------+--------+--------------------------+--------------------------+
|        ID        | STATUS  |  NAME  |        PEER ADDRS        |       CLIENT ADDRS       |
+------------------+---------+--------+--------------------------+--------------------------+
|   db5a2c24a8bf7a | started | etcd-2 | https://10.0.19.129:2380 | https://10.0.19.129:2379 |
| 318ff80236db7a1e | started | etcd-1 | https://10.0.19.127:2380 | https://10.0.19.127:2379 |
| ee6138446f578296 | started | etcd-3 | https://10.0.19.130:2380 | https://10.0.19.130:2379 |
+------------------+---------+--------+--------------------------+--------------------------+
#list 查看集群是否健康
[root@k8s-master1 etcd]# /opt/etcd/bin/etcdctl --ca-file=/opt/etcd/ssl/ca.pem --cert-file=/opt/etcd/ssl/server.pem --key-file=/opt/etcd/ssl/server-key.pem --endpoints="https://10.0.19.127:2379,https://10.0.19.129:2379,https://10.0.19.130:2379" cluster-health
member db5a2c24a8bf7a is healthy: got healthy result from https://10.0.19.129:2379
member 318ff80236db7a1e is healthy: got healthy result from https://10.0.19.127:2379
member ee6138446f578296 is healthy: got healthy result from https://10.0.19.130:2379
cluster is healthy

5、修改所有etcd配置并重

[root@k8s-master1 etcd]# cat /opt/etcd/cfg/etcd.conf|grep ETCD_INITIAL_CLUSTER
ETCD_INITIAL_CLUSTER="etcd-2=https://10.0.19.129:2380,etcd-1=https://10.0.19.127:2380,etcd-3=https://10.0.19.130:2380"
[root@k8s-master1 etcd]#  systemctl restart etcd
#list 查看集群是否健康
[root@k8s-master1 etcd]# /opt/etcd/bin/etcdctl --ca-file=/opt/etcd/ssl/ca.pem --cert-file=/opt/etcd/ssl/server.pem --key-file=/opt/etcd/ssl/server-key.pem --endpoints="https://10.0.19.127:2379,https://10.0.19.129:2379,https://10.0.19.130:2379" cluster-health
member db5a2c24a8bf7a is healthy: got healthy result from https://10.0.19.129:2379
member 318ff80236db7a1e is healthy: got healthy result from https://10.0.19.127:2379
member ee6138446f578296 is healthy: got healthy result from https://10.0.19.130:2379
cluster is healthy

etcd下线成功，需要把它配置到k8s

二、修改kube-apiserver服务连接的etcd地址重启

[root@k8s-master1 etcd]# cat /opt/kubernetes/cfg/*|grep 2379
--etcd-servers=https://10.0.19.127:2379,https://10.0.19.129:2379,https://10.0.19.130:2379 \
[root@k8s-master1 etcd]# systemctl restart kube-apiserver